DESCRIPTION: (Applicant's Abstract) Social network studies of HIV transmission and partner notification activities are crucial for understanding and preventing transmission of HIV and other infections. Both efforts involve the elicitation of HIV risk network data (i.e., names of and other information about specific sexual and drug injection partners). The success of these efforts in providing insight on and controlling the spread of HIV depends heavily on the completeness, reliability, and validity of these self-reported, recall-based data. Research suggests that the recall of sexual and injection partners is substantially incomplete. The overall goal of this small grant project is to improve interviewing procedures and techniques to maximize the completeness, reliability, and validity of recall-based data on sexual and drug injection partners in both research and partner notification contexts. This project focuses on the interviewing situation, including interviewer characteristics and interviewer behaviors, as it influences the quality and quantity of HIV risk network data elicited. The project consists of three main activities: 1) examination of interviewer effects on the completeness and reliability of HIV risk network data in secondary analysis of three extensive data sets (from two research studies and one data base of partner notification records); 2) qualitative description of the interviewing techniques used by veteran Disease Intervention Specialists (DIS) in their partner notification interviews; and 3) a randomized controlled trial of supplementary interviewing techniques and procedures, based in part on the results of the first two activities, for enhancing recall of sexual and drug injection partners in a partner notification context.